Civil society groups raise concerns over proposed changes to EU AI Act
More than 40 organisations warn that revisions under the AI Omnibus package could weaken safeguards and reduce transparency in the EU’s artificial intelligence framework.
ARTICLE 19 and over 40 civil society organisations have expressed concerns about proposed amendments to the European Union’s Artificial Intelligence Act under the so-called AI Omnibus package.
The groups argue that the changes could reduce protections for individuals affected by high-risk AI systems. Areas highlighted include biometric identification and the use of AI in sectors such as education, where safeguards are considered particularly important.
According to the organisations, the proposed revisions may weaken existing requirements and leave individuals with fewer protections against potential harms.
The statement also raises concerns about the legislative process. It notes that the European Commission did not carry out standard steps such as impact assessments or public consultations, which are typically used to evaluate potential effects and gather stakeholder input.
In response, ARTICLE 19 calls on EU institutions to maintain key elements of the original framework, including transparency obligations and oversight mechanisms.
The discussion reflects differing positions among stakeholders. While civil society groups emphasise the need for strong safeguards, some industry representatives have called for more flexible rules, arguing that this would support innovation and competitiveness.
